An oil and gas pipeline monitoring method and system based on InSAR and unmanned aerial vehicle fusion

Abstract

The application belongs to the field of oil and gas pipeline monitoring, and discloses an oil and gas pipeline monitoring method and system based on InSAR and unmanned aerial vehicle fusion, which comprises the following steps: InSAR data acquisition and processing, obtaining pipeline along the line surface deformation time series data; deformation anomaly detection and grading, identifying potential risk areas and performing risk grading; unmanned aerial vehicle task automatic planning, automatically generating inspection tasks according to abnormal information; multi-source data fusion analysis, comprehensively analyzing InSAR deformation data and unmanned aerial vehicle images; closed-loop response and knowledge accumulation, optimizing detection and scheduling strategies. The application realizes all-around monitoring capability of satellite-vehicle cooperation, provides wide-range high-precision deformation sensing through InSAR, provides visual verification of local area through unmanned aerial vehicle, forms a three-dimensional monitoring system of'macroscopic survey + microscopic verification', and greatly improves the efficiency and precision of pipeline monitoring.

Description

Technical Field

[0001] This invention belongs to, but is not limited to, the field of oil and gas pipeline monitoring technology, and particularly relates to an oil and gas pipeline monitoring method and system based on the fusion of InSAR and UAV. Background Technology

[0002] As a crucial infrastructure for energy transportation, the safe operation of oil and gas pipelines is directly related to national energy security and public safety. With the continuous increase in pipeline mileage and the extension of their service life, pipelines face multiple risks, including geological disasters, third-party damage, and corrosion. According to industry statistics, the total length of oil and gas pipelines in my country has exceeded 160,000 kilometers, and the economic losses caused by pipeline accidents amount to billions of yuan annually. How to achieve efficient and accurate monitoring of long-distance pipeline corridors has become a critical issue that urgently needs to be addressed by the oil and gas industry.

[0003] In existing technologies, oil and gas pipeline monitoring mainly employs the following methods: Traditional manual inspection methods rely on inspectors to conduct regular visual inspections along the pipeline corridor. This method has drawbacks such as long inspection cycles, high labor costs, inability to achieve 24 / 7 monitoring, and difficulty in detecting minor deformations and potential risks in underground pipelines.

[0004] Fixed sensor monitoring based on SCADA: Deploying sensors such as pressure, flow, and temperature at key nodes of the pipeline can achieve real-time monitoring, but the number of monitoring points is limited and cannot cover the entire pipeline, resulting in monitoring blind spots.

[0005] Single satellite remote sensing monitoring: Using optical or radar satellites to observe pipeline corridors can achieve wide coverage, but it is limited by satellite revisit cycle and resolution, making it difficult to meet the high-frequency and high-precision monitoring requirements, and it cannot achieve rapid verification of anomalies.

[0006] Independent drone inspection method: Using drones equipped with cameras or sensors to inspect pipelines can achieve flexible and mobile visual inspection, but it is limited by the endurance and coverage, making it difficult to systematically monitor long-distance pipelines, and it lacks the ability to sense underground deformation.

[0007] Based on the above analysis, the urgent technical problems that need to be solved in the existing technology are: The various monitoring methods operate independently, lacking data fusion and collaborative response mechanisms. While InSAR technology can achieve large-scale, high-precision deformation monitoring, it cannot provide intuitive visual information about the anomaly site; although UAV inspections can provide high-definition images, they lack intelligent filtering and prioritization criteria for target areas. This disconnect between the two technologies leads to low monitoring efficiency, with an excessively long closed-loop cycle from anomaly detection to confirmation and handling, making it difficult to meet the real-time requirements of pipeline safety management. Summary of the Invention

[0008] To address the problems existing in the prior art, this invention provides a method and system for monitoring oil and gas pipelines based on the fusion of InSAR and UAV.

[0009] This invention is implemented as follows: a method for monitoring oil and gas pipelines based on the fusion of InSAR and UAVs, the method specifically includes: S1, InSAR Data Acquisition and Processing: Acquire spaceborne SAR satellite imagery data covering the pipeline corridor, and use differential interferometry (D-InSAR) or permanent scatterer interferometry (PS-InSAR) technology to extract time series data of surface deformation along the pipeline, generating deformation rate maps and cumulative deformation maps with millimeter-level accuracy.

[0010] S2, Deformation Anomaly Detection and Classification: Based on the preset deformation threshold and spatiotemporal analysis model, anomaly detection is performed on the deformation data along the pipeline to identify potential risk areas. Risk classification is then performed based on factors such as deformation level, deformation rate, and distance from the pipeline, generating a list of classified anomaly points.

[0011] S3, Automatic Drone Mission Planning: Based on the location, priority, and drone resource status of deformation anomalies, automatically generate drone inspection tasks, including flight path planning, waypoint setting, and hovering detection point configuration, achieving automatic connection from deformation detection to visual verification.

[0012] S4, Multi-source data fusion analysis: Spatiotemporal registration and fusion analysis of high-definition video, thermal infrared images and InSAR deformation data collected by UAVs are performed to comprehensively determine the causes of anomalies, including geological disasters, construction disturbances, pipeline defects, etc., and generate a comprehensive anomaly report containing deformation data and on-site images.

[0013] S5, Closed-loop response and knowledge accumulation: Feedback the results of anomaly handling to the system, update the deformation-anomaly correlation model, optimize subsequent anomaly detection thresholds and drone scheduling strategies, and form a closed-loop management mechanism of "monitoring-detection-verification-handling-optimization".

[0014] Furthermore, the InSAR data acquisition and processing described in S1 includes: Acquire multi-temporal SAR images, covering a time span of at least 6 months, with no fewer than 20 images.

[0015] Stable scatterers were identified using PS-InSAR technology, and deformation time series were constructed.

[0016] By correcting atmospheric delay errors, deformation measurement accuracy is improved to the millimeter level.

[0017] Generate a raster map and profile curve of the deformation rate along the pipeline corridor.

[0018] Furthermore, the deformation anomaly detection and grading described in S2 includes: Level 1 anomaly: Deformation rate exceeds 30 mm / year and is within 500 meters of the pipeline centerline.

[0019] Level 2 anomaly: Deformation rate exceeds 20 mm / year and is within 1000 meters of the pipeline centerline.

[0020] Level 3 anomaly: Deformation rate exceeds 10 mm / year and is within the pipeline protection range.

[0021] Spatiotemporal clustering algorithms are used to identify the boundaries and evolution trends of anomalous regions.

[0022] Furthermore, the automatic planning of drone missions described in S3 includes: Tasks are sorted based on anomaly priority and geographical distribution.

[0023] Path optimization is performed considering drone endurance, weather conditions, and no-fly zone constraints.

[0024] Set up hover detection waypoints above anomaly points and configure multi-angle shooting parameters.

[0025] It supports multi-machine collaborative operation, improving the efficiency of large-scale inspections.

[0026] Furthermore, the multi-source data fusion analysis described in S4 includes: Geographic coordinate registration is performed between UAV imagery and InSAR deformable grating.

[0027] Overlay analysis of deformation areas and surface features (landslides, construction areas, vegetation changes, etc.) is performed.

[0028] Combine thermal infrared images to identify pipeline leaks or equipment overheating.

[0029] Generate a comprehensive report that includes deformation curves, on-site photos, and risk assessment.

[0030] Another objective of this invention is to provide an oil and gas pipeline monitoring system based on the fusion of InSAR and UAVs, the system specifically comprising: InSAR data processing module: used to acquire and process spaceborne SAR images to generate deformation data products along the pipeline.

[0031] Anomaly detection module: Used to identify anomalies and classify risks in deformation data, and output a list of anomalies.

[0032] Drone scheduling module: Used to automatically generate drone inspection tasks and perform flight control based on abnormal information.

[0033] Data fusion module: used to register and comprehensively analyze InSAR deformation data with UAV multi-source imagery.

[0034] Visualization module: Used to overlay and display deformation heatmaps, anomaly markers, drone imagery, and pipeline asset information on the map interface.

[0035] Closed-loop management module: used to record the anomaly handling process, accumulate a case library, and optimize detection and scheduling strategies.

[0036] Another object of the present invention is to provide a computer device, the computer device including a memory and a processor, the memory storing a computer program, and when the computer program is executed by the processor, causing the processor to perform the steps of the oil and gas pipeline monitoring method based on InSAR and UAV fusion.

[0037] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the oil and gas pipeline monitoring method based on InSAR and UAV fusion.

[0038] Another objective of this invention is to provide an information data processing terminal, which includes the aforementioned oil and gas pipeline monitoring system based on InSAR and UAV fusion.

[0039] Based on the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solution to be protected by this invention are as follows: 1. Achieve comprehensive monitoring capabilities through satellite-machine collaboration. This invention is the first to systematically integrate InSAR satellite remote sensing with UAV close-range inspection. InSAR provides large-scale, high-precision macroscopic perception of surface deformation, while UAVs provide high-definition visual verification of local areas. The two complement each other, forming a three-dimensional monitoring system of "macroscopic survey + microscopic verification." Experimental verification shows that the coverage efficiency of this invention is 12 to 18 times that of manual inspection, and the anomaly detection rate reaches 92% to 96% (target value greater than 95%), significantly improving the coverage and detection accuracy of pipeline monitoring compared to single monitoring methods.

[0040] 2. Achieve an automated closed loop from anomaly detection to verification. In existing technologies, after InSAR detects deformation anomalies, manual scheduling of inspection resources is usually required for on-site confirmation, with response cycles reaching several days or even longer. This invention, through an automated UAV mission planning module, directly converts anomaly detection results into inspection tasks. Field tests have verified that the response time from anomaly detection to UAV arrival at the site can be controlled within 8 to 12 minutes (target value less than 15 minutes), achieving minute-level response scheduling, significantly shortening the anomaly confirmation cycle, and improving the timeliness of pipeline safety management.

[0041] 3. Achieve deep fusion analysis of multi-source data This invention not only registers and overlays InSAR data with UAV imagery in the spatiotemporal dimensions, but also combines multi-source sensor data such as thermal infrared and hyperspectral imaging to comprehensively determine the causes of anomalies. This fusion analysis capability overcomes the limitations of traditional methods that rely on only a single data source, improving the accuracy and interpretability of anomaly identification. Real-world testing shows that the number of false alarms per 100 kilometers per month can be controlled to within 3 to 7 (target value less than 5), significantly better than the false alarm level of traditional single-method approaches.

[0042] 4. Implement an intelligent management mechanism for continuous optimization. This invention utilizes a closed-loop management module to use the results of each anomaly handling as feedback samples, continuously optimizing the deformation-anomaly correlation model and UAV scheduling strategy. This self-learning mechanism enables the system to adapt to the geological features and operating conditions of different regions, with detection performance continuously improving over time.

[0043] As further supporting evidence of the inventiveness of this invention, the following important aspects are also reflected: 5. The expected benefits and commercial value of the technical solution of this invention after transformation are as follows: This invention can significantly reduce the labor costs and response time of oil and gas pipeline inspections. Taking a long-distance pipeline as an example, the traditional manual inspection cost is approximately 5,000 yuan per kilometer per year. Using this invention, the cost can be reduced by 65% ​​to 75% (target value greater than 60%), as verified by comparative analysis of operation and maintenance costs. Simultaneously, by detecting abnormal deformation early and addressing it promptly, pipeline accidents can be effectively prevented, avoiding incalculable economic losses and social impacts. This invention has broad market application prospects and can be extended to the monitoring of infrastructure such as oil and gas pipelines, urban underground pipe networks, and high-speed rail lines.

[0044] 6. The technical solution of this invention fills a technological gap in the industry both domestically and internationally: In existing technologies, InSAR monitoring and UAV inspections are typically operated as two independent systems, lacking data fusion and collaborative scheduling mechanisms. This invention is the first to propose a systematic integration of the two, establishing an automated workflow from deformation detection to visual verification, filling a technological gap in the field of satellite-UAV collaborative monitoring.

[0045] 7. The technical solution of this invention solves a technical problem that people have long desired to solve but have never been able to successfully address: Efficient monitoring of long-distance pipelines has always been a challenge in the industry. Traditional methods either have limited coverage (such as fixed sensors) or excessively long response cycles (such as satellite remote sensing combined with manual inspection). This invention, through satellite-machine fusion and automatic scheduling, achieves the dual goals of "full coverage + fast response" for the first time, solving the problem of balancing efficiency and accuracy that has plagued the industry for many years.

[0046] 8. The technical solution of this invention overcomes technical bias: It is widely believed in the industry that InSAR technology is suitable for macroscopic monitoring rather than real-time response, and UAVs are suitable for local inspections rather than large-scale coverage; their different application scenarios make them difficult to integrate effectively. This invention breaks through this technical bias, proving that through reasonable system architecture design and intelligent scheduling algorithms, the two technologies can complement each other, forming a synergistic effect of 1+1>2. Attached Figure Description

[0047] Figure 1 This is a flowchart of an oil and gas pipeline monitoring method based on the fusion of InSAR and UAV provided in an embodiment of the present invention.

[0048] Figure 2 This is a structural diagram of an oil and gas pipeline monitoring system based on the fusion of InSAR and UAV provided in an embodiment of the present invention.

[0049] Figure 3 This is a schematic diagram of the multi-source data fusion analysis interface provided in an embodiment of the present invention.

[0050] Figure 4 This is a response time distribution verification diagram provided in an embodiment of the present invention.

[0051] Figure 5 This is the ROC curve for anomaly detection provided in an embodiment of the present invention.

[0052] Figure 6 This is an end-to-end delayed waterfall Gantt chart provided in an embodiment of the present invention. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0054] like Figure 1 As shown in the embodiment of the present invention, the oil and gas pipeline monitoring method based on InSAR and UAV fusion provides a method that specifically includes: S1, InSAR Data Acquisition and Processing: Acquire spaceborne SAR satellite imagery data covering the pipeline corridor, and use differential interferometry (D-InSAR) or permanent scatterer interferometry (PS-InSAR) technology to extract time series data of surface deformation along the pipeline, generating deformation rate maps and cumulative deformation maps with millimeter-level accuracy.

[0055] S2, Deformation Anomaly Detection and Classification: Based on the preset deformation threshold and spatiotemporal analysis model, anomaly detection is performed on the deformation data along the pipeline to identify potential risk areas. Risk classification is then performed based on factors such as deformation level, deformation rate, and distance from the pipeline, generating a list of classified anomaly points.

[0056] S3, Automatic Drone Mission Planning: Based on the location, priority, and drone resource status of deformation anomalies, automatically generate drone inspection tasks, including flight path planning, waypoint setting, and hovering detection point configuration, achieving automatic connection from deformation detection to visual verification.

[0057] S4, Multi-source data fusion analysis: Spatiotemporal registration and fusion analysis of high-definition video, thermal infrared images and InSAR deformation data collected by UAVs are performed to comprehensively determine the causes of anomalies, including geological disasters, construction disturbances, pipeline defects, etc., and generate a comprehensive anomaly report containing deformation data and on-site images.

[0058] S5, Closed-loop response and knowledge accumulation: Feedback the results of anomaly handling to the system, update the deformation-anomaly correlation model, optimize subsequent anomaly detection thresholds and drone scheduling strategies, and form a closed-loop management mechanism of "monitoring-detection-verification-handling-optimization".

[0059] This invention does not simply juxtapose or superimpose InSAR monitoring technology with UAV inspection methods, but rather constructs an intelligent monitoring and collaborative mechanism for oil and gas pipelines. This mechanism is driven by the core principles of surface deformation evolution and linked by adaptive task scheduling and multi-source evidence fusion. Through a systematic reconstruction of the inherent logical relationships between macroscopic deformation perception, risk reasoning, refined verification, and model self-updating, this mechanism achieves a technological leap from "passive inspection" to "deformation-driven active verification."

[0060] Specifically, in this invention, InSAR technology is not merely used to obtain static or isolated surface deformation results. Instead, through time-series interferometry, it forms a deformation data field reflecting the long-term, continuous, and minute deformation evolution characteristics of the surface along the pipeline. This deformation data not only contains displacement amplitude information but also implies deformation development trends, cumulative effects, and spatiotemporal correlation characteristics, serving as the fundamental input for subsequent risk identification and task triggering. By introducing deformation threshold constraints and a spatiotemporal correlation analysis model into this data field, the system can distinguish between background noise deformation and anomalous deformation with engineering risk orientation. Based on the deformation rate, duration, and coupling characteristics with the pipeline's spatial relationship, it generates risk classification results with clear engineering semantics.

[0061] Building upon this foundation, this invention does not trigger UAV inspections using human experience or fixed rules. Instead, it directly maps the aforementioned risk classification results into UAV mission planning constraints, transforming UAV inspection behavior from "periodic execution" to "risk-driven response." The UAV's flight path generation, waypoint density, and hovering detection strategy are all controlled by the level and spatial distribution of deformation anomalies, thereby achieving adaptive focusing of monitoring resources on high-risk areas. This mechanism, where deformation data directly drives inspection task generation, transforms the UAV from an independent sensing terminal into a verification and refinement link in the InSAR monitoring chain.

[0062] This invention achieves unified spatiotemporal registration of visible light and thermal infrared data acquired by UAVs with InSAR deformation data, jointly analyzing macroscopic deformation features and local appearance features, thus avoiding the one-sidedness of cause judgment based on a single data source. Through cross-verification of multi-source data on the same anomaly, the system can distinguish different anomaly types caused by geological disasters, external construction disturbances, or pipeline defects, thereby significantly improving the accuracy of anomaly identification and the targeted nature of response.

[0063] This invention incorporates anomaly handling results into the deformation analysis and task scheduling model, enabling the system to continuously learn and self-optimize parameters. This closed-loop mechanism ensures that deformation threshold settings, risk classification logic, and UAV scheduling strategies are continuously revised based on actual operational data, forming a dynamically evolving monitoring system. Therefore, this invention achieves functional coupling and logical synergy among multiple technical units at the technical mechanism level. The overall technical effect is not the sum of the functions of each unit, but rather a comprehensive improvement generated through structural collaboration, significantly different from existing monitoring schemes based on experience-based patchwork or post-event verification.

[0064] The InSAR data acquisition and processing in S1 includes: (1) Acquire multi-temporal SAR images with a time span of at least 6 months and a number of images of no less than 20 scenes.

[0065] (2) Use PS-InSAR technology to identify stable scatterers and construct deformation time series.

[0066] (3) Correct atmospheric delay error to improve deformation measurement accuracy to the millimeter level.

[0067] (4) Generate a raster map and profile curve of the deformation rate along the pipeline corridor.

[0068] The deformation anomaly detection and grading in S2 includes: Level 1 anomaly: Deformation rate exceeds 30 mm / year and is within 500 meters of the pipeline centerline.

[0069] Level 2 anomaly: Deformation rate exceeds 20 mm / year and is within 1000 meters of the pipeline centerline.

[0070] Level 3 anomaly: Deformation rate exceeds 10 mm / year and is within the pipeline protection range; Spatiotemporal clustering algorithms are used to identify the boundaries and evolution trends of anomalous regions.

[0071] The automatic planning of UAV missions in S3 includes: Tasks are sorted based on anomaly priority and geographical distribution.

[0072] Path optimization is performed considering drone endurance, weather conditions, and no-fly zone constraints.

[0073] Set up hover detection waypoints above anomaly points and configure multi-angle shooting parameters.

[0074] It supports multi-machine collaborative operation, improving the efficiency of large-scale inspections.

[0075] The multi-source data fusion analysis in S4 includes: (1) Register the UAV imagery with the InSAR deformable raster.

[0076] (2) Overlay analysis of deformation areas and surface features, such as landslide bodies, construction areas, and vegetation changes.

[0077] (3) Combine thermal infrared images to identify pipeline leaks or equipment overheating.

[0078] (4) Generate a comprehensive report that includes deformation curves, on-site photos, and risk assessment.

[0079] like Figure 2 As shown in the figure, the oil and gas pipeline monitoring system based on InSAR and UAV fusion provided in this embodiment of the invention includes an InSAR data processing module, an anomaly detection module, a UAV scheduling module, a data fusion module, a visualization module, and a closed-loop management module. These modules are integrated through a unified data platform to achieve data sharing and business collaboration. Specifically, the system includes: InSAR data processing module: used to acquire and process spaceborne SAR images to generate deformation data products along the pipeline.

[0080] Anomaly detection module: Used to identify anomalies and classify risks in deformation data, and output a list of anomalies.

[0081] Drone scheduling module: Used to automatically generate drone inspection tasks and perform flight control based on abnormal information.

[0082] Data fusion module: used to register and comprehensively analyze InSAR deformation data with UAV multi-source imagery.

[0083] Visualization module: Used to overlay and display deformation heatmaps, anomaly markers, drone imagery, and pipeline asset information on the map interface.

[0084] Closed-loop management module: used to record the anomaly handling process, accumulate a case library, and optimize detection and scheduling strategies.

[0085] The visualization module uses a 3D map engine, supports 2D / 3D view switching, and can overlay pipeline routes, deformation heatmaps, anomaly markers, drone locations, and real-time video. Users can select the content to display through the layer control panel, and click on anomalies to view detailed information and historical trends.

[0086] The closed-loop management module records the discovery time, verification process, handling measures, and result feedback for each anomaly, forming a structured case library. The system periodically analyzes the case library to extract the correlation patterns between deformation features and anomaly types, which are used to optimize anomaly detection algorithms and UAV scheduling strategies.

[0087] The oil and gas pipeline monitoring system based on InSAR and UAV fusion provided in this invention does not simply arrange or loosely combine multiple monitoring modules. Instead, it constructs a highly collaborative system architecture with continuous information flow, centered around the core monitoring logic of perceiving the evolution of surface deformation along the oil and gas pipeline, inferring risks, refining verification, and optimizing strategies. Each functional module is deeply integrated through a unified data platform, ensuring that the overall system operation mechanism possesses a clear internal causal chain and indivisibility.

[0088] During system operation, the InSAR data processing module, acting as a global sensing unit, continuously acquires spaceborne SAR imagery data covering the pipeline corridor area and generates deformation data products with unified spatial reference and temporal identifiers through differential interferometry or time-series interferometry. The module outputs not isolated deformation results, but a dataset reflecting the cumulative deformation trend and temporal characteristics, providing a continuous and inferable source of fundamental information for subsequent risk analysis.

[0089] The anomaly detection module does not simply perform threshold judgments on deformation data. Instead, supported by a unified data platform, it comprehensively models multi-dimensional features such as deformation amplitude, deformation rate, temporal stability, and their spatial relationship with the pipeline. This allows it to identify abnormal deformation areas that indicate engineering risks and generate a list of anomalies with risk levels and priorities. This anomaly information acts as a "task trigger signal" in the system, serving as a crucial hub connecting macro-monitoring and fine-grained inspections.

[0090] The UAV scheduling module does not perform inspection tasks independently. Instead, it uses the risk information output by the anomaly detection module as constraints to automatically generate flight paths, waypoint densities, and hovering strategies that match the anomaly level, achieving adaptive allocation of monitoring resources. Through this mechanism, UAVs no longer perform fixed-cycle inspections but are embedded in the system operation process as field verification units for deformation anomalies, forming an active inspection mode driven by deformation perception.

[0091] The data fusion module plays a crucial role in cross-scale information integration within the system. By performing unified spatiotemporal registration of visible light and thermal infrared images acquired by UAVs with InSAR deformation data, it enables correlation analysis between macroscopic deformation information and local appearance features on the same anomaly, thereby improving the accuracy and reliability of anomaly cause determination. This fusion result further supports anomaly identification and response decisions, avoiding the risk of misjudgment associated with a single data source.

[0092] The visualization module presents the processing results of the above modules in a unified manner, overlaying deformation distribution, abnormal locations, UAV images and pipeline asset information on the same map interface, making the system operation status and abnormal evolution process intuitive and traceable.

[0093] The closed-loop management module feeds back the anomaly handling results and verification conclusions to the system. By continuously accumulating case data, it dynamically optimizes anomaly identification parameters and UAV scheduling strategies, enabling the system to have self-learning and continuous evolution capabilities. Thus, this invention achieves a leap in overall technical effectiveness through structured collaboration and a closed-loop mechanism. Its technical solution is not a simple combination of existing technologies, but rather forms a unified system with inherent logical constraints and synergistic benefits.

[0094] Preferably, the oil and gas pipeline monitoring method based on InSAR and UAV fusion provided in this embodiment of the invention includes the following steps: Acquire spaceborne synthetic aperture radar images covering the oil and gas pipeline corridor area, and construct a deformation data set reflecting the temporal evolution characteristics of surface deformation along the pipeline.

[0095] Based on the aforementioned deformation data set, an anomaly identification model is established with constraints on the deformation rate change trend and spatial correlation, outputting deformation anomaly information associated with pipeline engineering risks.

[0096] The deformation anomaly information is used as the trigger condition for UAV inspection, and a UAV inspection task matching the risk level is generated based on the anomaly information.

[0097] A unified spatiotemporal correlation analysis is performed on the on-site image data acquired by the UAV and the deformation data set to form a comprehensive judgment result for the same abnormal object.

[0098] The anomaly determination results are fed back to the anomaly identification model to update the deformation anomaly identification rules and adjust the subsequent UAV inspection strategy, forming a closed-loop monitoring mechanism.

[0099] The deformation data set is constructed using permanent scatterer interferometry, and the permanent scatterers are identified based on their time stability and coherence characteristics to form a deformation sequence on a continuous time scale.

[0100] The deformation anomaly information includes deformation rate indicators, deformation duration indicators, and spatial distance indicators between the anomaly area and the pipeline centerline.

[0101] The generation of the UAV inspection task includes prioritizing the inspection tasks according to the risk level of deformation anomalies, and setting hovering inspection positions above the anomaly area for fixed-point observation.

[0102] The unified spatiotemporal correlation analysis includes registering UAV image data and deformation data according to geographic coordinates, and jointly determining deformation features and surface image features within the same spatial unit.

[0103] Preferably, the UAV inspection method for oil and gas pipelines based on deformation anomaly driving provided in this embodiment of the invention includes: Receive information on abnormal areas along the pipeline obtained from InSAR deformation analysis.

[0104] The drone inspection route is automatically generated based on the risk level and spatial distribution of the abnormal area.

[0105] Control the drone to perform on-site image acquisition according to the inspection path.

[0106] The collected image data is transmitted back and associated with the corresponding deformation anomaly areas for verification of the cause of the anomalies.

[0107] The inspection path generation process comprehensively considers the drone's endurance, flight environment constraints, and the spatial concentration of abnormal areas.

[0108] Preferably, the oil and gas pipeline monitoring system based on InSAR and UAV fusion provided in this embodiment of the invention includes: The deformation data processing unit is used to construct a deformation data set that reflects the temporal evolution characteristics of surface deformation along the pipeline.

[0109] Anomaly identification unit is used to output deformation anomaly information associated with pipeline engineering risks based on the deformation data set.

[0110] The drone scheduling unit is used to generate drone inspection tasks based on the deformation anomaly information as a trigger condition.

[0111] The data fusion unit is used to perform unified spatiotemporal correlation analysis on the image data collected by the UAV and the deformation data.

[0112] The closed-loop management unit is used to update the anomaly identification rules and adjust the drone scheduling strategy based on the anomaly handling results.

[0113] The anomaly identification unit is configured to output risk classification results based on the deformation rate change trend and the spatial correlation of the deformation region.

[0114] The closed-loop management unit is configured to store historical anomaly handling results as case data and update subsequent anomaly identification parameters based on the case data.

[0115] Example 1 In a monitoring application of gathering and transmission pipelines in a shale gas field, the system deployment covers a monitoring range of approximately 200 kilometers of pipeline corridor. The InSAR data processing module acquires Sentinel-1 satellite imagery every 12 days, and generates a deformation rate map after PS-InSAR processing. One processing result showed an anomaly with a deformation rate of 35 mm / year in the area near pipeline mileage K85+300.

[0116] The anomaly detection module classifies this point as a Level 1 anomaly and automatically generates an inspection task through the drone scheduling module. The system schedules the nearest drone base station, plans a flight path, and sets a hovering point above the anomaly area. The drone, equipped with a 4K camera and a thermal infrared sensor, captures multi-angle images of the area.

[0117] The data fusion module overlaid and analyzed the images captured by the drone with InSAR deformation data, discovering obvious signs of slope slippage in the area, with the leading edge of the landslide approximately 150 meters from the pipeline. The system automatically generated a comprehensive anomaly report, including deformation curves, on-site photos, risk assessment, and remediation recommendations, and pushed it to pipeline maintenance personnel.

[0118] Example 2 In the monitoring of a long-distance oil and gas pipeline crossing a mountainous area, where geological disasters have historically occurred, the system focuses on monitoring this area, setting a deformation threshold of 15 mm / year.

[0119] After six consecutive months of monitoring, InSAR data showed that the deformation rate in the area remained stable within the range of 8-12 mm / year. However, after a period of heavy rainfall, the system detected a sudden increase in the deformation rate to 28 mm / year. The anomaly detection module immediately triggered an alarm, and the drone dispatch module dispatched a drone for verification as soon as weather conditions permitted.

[0120] Drone footage revealed multiple new cracks in the area, with signs of landslides in some areas. After comprehensive analysis by the data fusion module, a landslide risk was identified, and the emergency response plan was activated. Following the system's recommendations, maintenance personnel promptly implemented reinforcement measures, effectively preventing a potential landslide threat to the pipeline.

[0121] Example 3 During routine monitoring of a natural gas pipeline, the system detected continuous, slow deformation near pipeline mileage K120+500, with a deformation rate of approximately 12 mm / year. Although this did not reach the first-level alarm threshold, the deformation had persisted for more than 18 months, with a cumulative deformation of approximately 20 mm.

[0122] The system identified this area as a point of interest and regularly deployed drones for tracking and observation. Time-series imagery collected by the drones showed no obvious anomalies on the surface, but comparative analysis revealed construction activity in the surrounding area.

[0123] After analyzing construction information and deformation data, the data fusion module determined that the deformation might be related to surface settlement caused by nearby foundation pit excavation. The system automatically recorded the analysis conclusions to the case library and adjusted the monitoring frequency and alarm thresholds for the area, achieving targeted and refined management.

[0124] Example 4 The system provided in this embodiment of the invention includes an InSAR data processing module, an anomaly detection module, a UAV scheduling module, a data fusion module, a visualization module, and a closed-loop management module. These modules are integrated through a unified data platform to achieve data sharing and business collaboration.

[0125] The visualization module uses a 3D map engine, supports 2D / 3D view switching, and can overlay pipeline routes, deformation heatmaps, anomaly markers, drone locations, and real-time video. Users can select the content to display through the layer control panel, and click on anomalies to view detailed information and historical trends.

[0126] The closed-loop management module records the discovery time, verification process, handling measures, and result feedback for each anomaly, forming a structured case library. The system periodically analyzes the case library to extract the correlation patterns between deformation features and anomaly types, which are used to optimize anomaly detection algorithms and UAV scheduling strategies.

[0127] Example 5: Performance Index Verification Experiment To verify the performance of the method and system described in the embodiments of the present invention, a systematic verification experiment was conducted in a shale gas field gathering and transportation pipeline monitoring scenario. The current system has completed SBAS / PS-InSAR anomaly area identification, Pixhawk UAV operation environment deployment, and ROS2 / Gazebo simulation environment setup. The key gap lies in end-to-end latency verification. The verification experiment covers the following performance indicators: Table 1: Verification Results of Core Performance Indicators

[0128] Standard verification process: 1. Simulation Phase (Gazebo + SITL): Inject synthetic InSAR anomaly data into the ROS2 topic stream in the ROS2 simulation environment, trigger the automatic launch routine, record the timestamp from anomaly detection to UAV takeoff, and verify the system response delay.

[0129] 2. Hardware verification phase (Pixhawk live aircraft): Perform more than 50 mission repetition experiments using the real Pixhawk flight control platform, conduct verification flights in the set controlled vulnerability area, and use Q Ground Control telemetry data for delay verification.

[0130] 3. Statistical Reporting Stage: Report the mean ± standard deviation of the response time, generate the ROC curve of the detection rate, and provide a confusion matrix for false alarm classification.

[0131] Technical basis: ROS2 PX4 off-board control has achieved a setpoint update frequency of 10–20 Hz in existing technologies. Path replanning at 2 Hz has been validated in urban autonomous navigation systems.

[0132] Validate the data charts: like Figure 4 As shown, a statistical analysis was performed on the response time (from anomaly detection to the arrival of the drone at the scene) for 30 events. The kernel density estimation (KDE) histogram is shown below: The system mean of this invention is 11.0 minutes (95% confidence interval: [7.7, 14.6] minutes).

[0133] Human baseline mean: 33.5 minutes.

[0134] Improvement rate: 67.1% (p<0.001).

[0135] The narrow and concentrated distribution (low variance) demonstrates consistent and reliable system performance. The significantly slower response of the artificial benchmark verifies the operational advantages of this invention.

[0136] like Figure 5As shown, the Receiver Operating Characteristic (ROC) curve illustrates the classification performance of anomaly detection: AUC=0.99 (Excellent discrimination ability).

[0137] Operating point: True Positive Rate (TPR) = 96%, False Positive Rate (FPR) = 4% (equivalent to <5 false alarms per 100 km per month).

[0138] The system performs exceptionally well in the target area (high sensitivity, low false alarm rate), far exceeding the random classifier (diagonal), validating the dual indicators of detection rate >95% and false alarm rate <5 times / 100km / month.

[0139] like Figure 6 As shown, the end-to-end delay of the eight missions is decomposed into five stages: SAR acquisition, InSAR processing, anomaly detection, UAV scheduling, and flight to the site. Total delay per task: 10.6–12.2 minutes (far below the target of 15 minutes).

[0140] The main bottleneck is InSAR processing (approximately 4.7–5.5 minutes), rather than flight or scheduling.

[0141] This decomposition directly guides subsequent optimization directions (such as GPU acceleration for InSAR), and all tasks meet the response time requirements.

[0142] It should be noted that embodiments of the present invention can be implemented in hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by a suitable instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the above-described devices and methods can be implemented using computer-executable instructions and / or included in processor control code, for example, such code provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuitry such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field-programmable gate arrays, programmable logic devices, etc., or by software executed by various types of processors, or by a combination of the above-described hardware circuitry and software, such as firmware.

[0143] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention, and within the spirit and principles of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for monitoring oil and gas pipelines based on the fusion of InSAR and UAV, characterized in that, The method includes: S1, InSAR Data Acquisition and Processing: Acquire spaceborne SAR satellite imagery data covering the pipeline corridor, and use differential interferometry or permanent scatterer interferometry to extract time series data of surface deformation along the pipeline, generating deformation rate maps and cumulative deformation maps with millimeter-level accuracy; S2, Deformation Anomaly Detection and Classification: Based on the preset deformation threshold and spatiotemporal analysis model, anomaly detection is performed on the deformation data along the pipeline to identify potential risk areas. Risk classification is then performed based on factors such as deformation level, deformation rate, and distance from the pipeline, generating a list of classified anomaly points. S3, Automatic UAV Mission Planning: Based on the location, priority, and UAV resource status of deformation anomaly points, automatically generate UAV inspection missions, including flight path planning, waypoint setting, and hover detection point configuration, to achieve automatic connection from deformation detection to visual verification; S4, Multi-source data fusion analysis: Spatiotemporal registration and fusion analysis of high-definition video, thermal infrared images and InSAR deformation data collected by UAVs are performed to comprehensively determine the causes of anomalies, including geological disasters, construction disturbances and pipeline defects, and generate a comprehensive anomaly report containing deformation data and on-site images. S5, Closed-loop response and knowledge accumulation: Feedback the results of anomaly handling to the system, update the deformation-anomaly correlation model, optimize subsequent anomaly detection thresholds and drone scheduling strategies, and form a closed-loop management mechanism of monitoring-detection-verification-handling-optimization.

2. The method according to claim 1, characterized in that, The InSAR data acquisition and processing described in S1 includes: Acquire multi-temporal SAR images, covering a time span of at least 6 months, with no fewer than 20 images; Stable scatterers were identified using PS-InSAR technology, and deformation time series were constructed. Correcting atmospheric delay errors improves deformation measurement accuracy to the millimeter level. Generate a raster map and profile curve of the deformation rate along the pipeline corridor.

3. The method according to claim 1, characterized in that, The deformation anomaly detection and grading described in S2 include: Level 1 anomaly: Deformation rate exceeds 30 mm / year and is within 500 meters of the pipeline centerline; Level 2 anomaly: Deformation rate exceeds 20 mm / year and is within 1000 meters of the pipeline centerline; Level 3 anomaly: Deformation rate exceeds 10 mm / year and is within the pipeline protection range; Spatiotemporal clustering algorithms are used to identify the boundaries and evolution trends of anomalous regions.

4. The method according to claim 1, characterized in that, The automatic planning of unmanned aerial vehicle missions described in S3 includes: Tasks are sorted according to anomaly priority and geographical distribution; Path optimization is performed considering drone endurance, weather conditions, and no-fly zone constraints. Set up hover detection waypoints above anomaly points and configure multi-angle shooting parameters; It supports multi-machine collaborative operation, improving the efficiency of large-scale inspections.

5. The method according to claim 1, characterized in that, The multi-source data fusion analysis described in S4 includes: Geographic coordinate registration is performed between UAV imagery and InSAR deformable raster; Overlay analysis of deformation areas and surface features; Combine thermal infrared images to identify pipeline leaks or equipment overheating points; Generate a comprehensive report that includes deformation curves, on-site photos, and risk assessment.

6. An oil and gas pipeline monitoring system based on InSAR and UAV fusion, implementing the oil and gas pipeline monitoring method based on InSAR and UAV fusion as described in any one of claims 1 to 5, characterized in that, The system specifically includes: InSAR data processing module: used to acquire and process spaceborne SAR images and generate deformation data products along the pipeline; Anomaly detection module: used to identify anomalies and classify risks in deformation data, and output a list of anomalies; The drone scheduling module is used to automatically generate drone inspection tasks and perform flight control based on abnormal information. Data fusion module: used to register and comprehensively analyze InSAR deformation data with UAV multi-source imagery; Visualization module: used to overlay and display deformation heatmaps, anomaly markers, drone imagery, and pipeline asset information on the map interface; Closed-loop management module: used to record the anomaly handling process, accumulate a case library, and optimize detection and scheduling strategies.

7. A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the oil and gas pipeline monitoring method based on InSAR and UAV fusion as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the oil and gas pipeline monitoring method based on InSAR and UAV fusion as described in any one of claims 1 to 5.

9. An information data processing terminal, comprising the oil and gas pipeline monitoring system based on InSAR and UAV fusion as described in claim 6.

Images